ORIGINAL ARTICLE Open Access

Identification of a cold-tolerant locus in rice(Oryza sativa L.) using bulked segregantanalysis with a next-generation sequencingstrategyJian Sun1†, Luomiao Yang1†, Jingguo Wang1, Hualong Liu1, Hongliang Zheng1, Dongwei Xie2, Minghui Zhang3,Mingfang Feng3, Yan Jia1, Hongwei Zhao1 and Detang Zou1*

Abstract

Background: Cold stress can cause serious abiotic damage that limits the growth, development and yield of rice.Cold tolerance during the booting stage of rice is a key factor that can guarantee a high and stable yield undercold stress. The cold tolerance of rice is controlled by quantitative trait loci (QTLs). Based on the complex geneticbasis of cold tolerance in rice, additional efforts are needed to detect reliable QTLs and identify candidate genes. Inthis study, recombinant inbred lines (RILs) derived from a cross between a cold sensitive variety, Dongnong422, andstrongly cold-tolerant variety, Kongyu131, were used to screen for cold-tolerant loci at the booting stage of rice.

Results: A novel major QTL, qPSST6, controlling the percent seed set under cold water treatment (PSST) under thefield conditions of 17 °C cold water irrigation was located on the 28.4 cM interval on chromosome 6. Using thecombination of bulked-segregant analysis (BSA) and next-generation sequencing (NGS) technology (Seq-BSA), a 1.81 Mb region that contains 269 predicted genes on chromosome 6 was identified as the candidate region ofqPSST6. Two genes, LOC_Os06g39740 and LOC_Os06g39750, were annotated as “response to cold” by gene ontology(GO) analysis. qRT-PCR analysis revealed that LOC_Os06g39750 was strongly induced by cold stress. Haplotypeanalysis also demonstrate a key role of LOC_Os06g39750 in regulating the PSST of rice, suggesting that it was thecandidate gene of qPSST6.

Conclusions: The information obtained in this study is useful for gene cloning of qPSST6 and for breedingcold-tolerant varieties of rice using marker assisted selection (MAS).

Keywords: Oryza sativa L, Cold tolerance, QTL, Seq-BSA, Candidate gene

BackgroundRice, a staple food crop cultivated worldwide, feeds morethan half of the world’s population. Rice originates fromtropical and subtropical regions and is more sensitive tolow temperature than other crops originating in temper-ate zones (Wang et al. 2016). In high-latitude or high-altitude regions of Asia, Europe, America and other ricecultivation areas of the world, the temperature is not

consistently high enough for the growth of rice. Remark-ably, low temperatures that occur frequently during thereproductive stage of rice can cause a fatal yield loss(Jena et al. 2012; Shirasawa et al. 2012). Male sterility,arising from low temperature (lower than 19 °C) duringthe period of microspore development at the booting stage,is the key reason for the reduction of percent seed set andthe resulting loss in yield (Satake and Hayase 1970). There-fore, the breeding of cold-tolerant varieties at the bootingstage is an effective method to maintain high and stableyields in rice cultivation regions.The cold tolerance of rice is a complex trait, and many

methods have been established to evaluate and select

* Correspondence: zoudt@sina.cnJian Sun and Luomiao Yang contributed equally to this work.†Equal contributors1College of Agriculture, Northeast Agricultural University, Harbin 150030,ChinaFull list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made.

Sun et al. Rice (2018) 11:24 https://doi.org/10.1186/s12284-018-0218-1

cold-tolerant varieties of rice (Zhang et al. 2014). Acold-water irrigation system has been developed as a re-liable identified method of determining cold tolerance atthe booting stage of rice. Rice plants are maintained in acold deep-water irrigated pool during the entire bootingstage, and the spikelet fertility was used to examine thecold tolerance of rice varieties (Shirasawa et al. 2012).This method exposes rice plants in field growth condi-tions, and its accurate evaluation results are still widelyused for selecting cold-tolerant lines and developing cold-tolerant rice varieties (Matsunaga 2005; Jia et al. 2015).Many studies agree that the cold tolerance of rice is

controlled by quantitative trait loci (QTLs) (Andaya andMackill 2003a; Zhang et al. 2005; Suh et al. 2010; Zhanget al. 2014). QTL mapping is the main approach to exca-vate and clone cold-tolerant related genes in rice. At thebooting stage of rice, a number of QTLs about cold tol-erance have been reported, including qCTB3 (Andayaand Mackill 2003b), qPSST-3 (Suh et al. 2010) andqLTB3 (Shirasawa et al. 2012) on chromosome 3, Ctb1(Saito et al. 2004), Ctb2 (Saito et al. 2001), qCTB-4-1,qCTB-4-2 (Xu et al. 2008), and CTB4a (Zhang et al. 2017)on chromosome 4, and qPSST-7 (Suh et al. 2010) andqCTB7 (Zhou et al. 2010) on chromosome 7. For the dif-ferent rice materials and research backgrounds, QTLs forcold tolerance at the booting stage were mapped at thevarious locations on different chromosomes. However,the above QTLs identified by the biparental crosslinkage mapping method are labor and time intensiveto map the genotypes a large number of individualsin the segregated population and to finely map thetarget QTL (Salvi and Tuberosa 2005).A bulked segregant analysis (BSA) is a simple and rapid

approach to identify molecular markers tightly linked tothe target genes or QTLs (Michelmore et al. 1991). Withthe rapid development of next-generation sequencing(NGS) technology, the combination of the BSA with NGSstrategy (Seq-BSA) is becoming a widely used approach inthe mapping of major QTLs and gene identification(Wenger et al. 2010; Takagi et al. 2013). This strategyhas been demonstrated in many plants, such as Arabi-dopsis (Schneeberger et al. 2009), rice (Abe et al. 2012;Wambugu et al. 2017), soybean (Song et al. 2017),wheat (Trick et al. 2012), pigeon pea (Singh et al. 2016), sunflower (Livaja et al. 2013) and sorghum (Han et al.2015), and it has identified some QTLs and genes forimportant traits. In the present study, two strategies,traditional QTL mapping and Seq-BSA were employedto identify the genes for cold tolerance at the bootingstage in recombinant inbred lines (RILs) derived from across between a cold sensitive variety, Dongnong422,and a strongly cold-tolerant variety, Kongyu131. Theresults contribute to the understanding of the geneticbases for cold tolerance at the booting stage, and future

cloning of the candidate gene will facilitate the molecu-lar breeding of cold tolerance in rice.

MethodsPlant materialsTwo japonica varieties, “Dongnong422” (DN422) and“Kongyu131 (KY131)” were used as parental lines to de-velop the RIL population. DN422 is a cold-sensitive var-iety that was obtained from Northeast AgricultureUniversity. KY131, a strongly cold-tolerant variety, iswidely cultivated in the northeast region of China. Themapping populations of 190 F7, F8 and F9 RILs wereproduced by single seed descent (SSD) from an F2 popu-lation of a cross between DN422 and KY131.

Cold tolerance evaluationAn evaluation of cold tolerance was performed on theexperimental farm of Northeast Agricultural University,Harbin, China (47°98′N, 128°08′E) in 2014, 2015 and2016. The parents and RIL populations were grown in arandomized block design with two replications of doublerow plots, a 2-m row length, a 30-cm row spacing and a10-cm hill spacing. The 190 RILs were divided into threegroups, early maturing, middle-maturing and late-maturing groups, according to their heading dates. DN422belongs to the late-maturing group, and KY131 belongs tothe early maturing group. For the evaluation of cold toler-ance, each group was irrigated with 17 °C water in an irri-gated pool (25 m × 5 m) independently from the panicleinitiation stage (approximately 35 days after transplantingfor the early maturing group, approximately 40 days aftertransplanting for the middle-maturing group, and ap-proximately 45 days after transplanting for the late-maturing group) to the full heading stage. When the aur-icle of the flag leaf is approximately 5 cm below the auricleof the penultimate leaf on each plant, the pollen shouldhave undergone meiosis, which was indicative of panicleinitiation stage (Satake and Hayase 1970). Water at 17 °Cwas prepared in a storage pool with a cool and warmwater mixture measured by the temperature sensors.Flood irrigation was performed from the inlet to the outlet(5 m) of the irrigated pool. The depth of the irrigatedwater was 18–20 cm, and the irrigated time was from 8 a.m. to 16 p.m. every day.The percent seed set under the cold-water treatment

(PSST), the ratio of the number of fertile seeds in thenumber of total seeds, was used as the index of cold tol-erance. A basic statistical analysis was implemented bythe SPSS16.0 software (SPSS Inc., Chicago, IL, USA).The mean data of PSST in RILs over the two replica-tions was used for QTL analysis, and the extreme cold-tolerant and sensitive plants in the RIL population usedfor Seq-BSA were selected according to the mean dataof F7, F8 and F9 RILs.

Sun et al. Rice (2018) 11:24 Page 2 of 12

QTL analysisOne hundred eighty-five polymorphic SSR markers be-tween DN422 and KY131 covering the rice genome wereused for genotyping the RIL population. A PCR was per-formed according to the procedure of Chen et al. (1997),and the PCR products were then separated on a 6%polyacrylamide gel, followed by silver staining.The MAP function of QTL IciMapping 3.2 (Wang et

al. 2012) was used to construct the genetic linkage mapof the RIL population, and the Kosambi’s mapping func-tion was used to calculate the genetic distances. QTLswere detected using the inclusive composite intervalmapping (ICIM) module of QTL IciMapping 3.2. Thethreshold of the LOD score for declaring the presence ofa significant QTL was determined by the permutationtest with 1000 repetitions at P < 0.05. The QTL wasnamed according to the trait and its chromosomelocation.

Construction of segregating pools and sequencingFor Seq-BSA, two DNA pools were developed by select-ing the extreme cold-tolerant and extreme cold-sensitiveindividuals according to the PSST of the RIL populationin the range from 0.50–0.95. The tolerant pool (T-pool)was made by mixing equal amounts of DNA from 20 ex-treme cold-tolerant RILs with a PSST above 0.90, andthe sensitive pool (S-pool) was made by mixing equalamounts of DNA from 20 extreme cold-sensitive RILswith a PSST below 0.63 (Additional file 2: Table S1).The DNA isolated from DN422 and KY131 and the twoDNA pools were prepared for sequencing.Libraries for all the DNA pools were prepared accord-

ing to the Illumina TruSeq DNA sample Preparation v2Guide. The DNA libraries were sequenced on IlluminaMiSeq platform using MiSeq Reagent Kit v2 (500 cycles)(Illumina Inc., San Diego, CA, USA). The short readsfrom both parents and the two DNA pools were alignedto Nipponbare reference genome (IRGSP 2005) usingthe BWA software (Li and Durbin 2009). Reads of the T-pool and S-pool were separately aligned to KY131 andDN422 consensus sequence reads to call SNPs with theSAM tools software (Li and Durbin 2009).

Analysis of the Seq-BSA dataAccording to the locating results of clean reads among thereference genome, duplicate reads were removed usingthe Picard tool (http://sourceforge.net/projects/picard/),and GATK software (McKenna et al. 2010) was used toperform the local realignment and base recalibration toinsure the accuracy of the SNP detecting. The SNP locibetween the test samples and reference genome were ob-tained using the GATK software according to the bestpractices on the GATK website (https://www.broadinsti-tute.org/gatk/guide/best-practices.php). All the SNP loci

between the test samples were summarized according tothe alignment results of test samples and the referencegenome.The Euclidean distance (ED) and SNP-index were calcu-

lated to identify the candidate regions of the genome associ-ated with PSST. The ED algorithm is a method of searchingmarkers with significant differences between the pools ac-cording to the sequencing data and evaluating the associatedregions between markers and traits (Hill et al. 2013).The calculation formula of the ED algorithm was as

follows:

ED ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

Aaa−Aabð Þ2 þ Caa−Cabð Þ2 þ Gaa−Gabð Þ2 þ Taa−Tabð Þ2q

where Aaa, Caa, Gaa and Taa represent the frequency ofbases A, C, G and T in the T-pool, respectively. Aab, Cab,Gab and Tab represent the frequency of bases A, C, Gand T in the S-pool, respectively. The depth of each basein different pools and the ED value of each SNP lociwere calculated. To eliminate the background noise, theED value was powered, and the ED5 was used as theassociated value (Hill et al. 2013).The SNP-index association analysis is a method used

to calculate genotype frequency differences between twopools (Takagi et al. 2013; Fekih et al. 2013). A SNP-index is the proportion of reads harboring the SNP thatare different from the reference sequence. The Δ(SNP-index) of each locus was calculated by subtraction of theSNP-index of the T-pool from that of the S-pool. SNP-index = 0 if the entire short reads contain genomic frag-ments from DN422; SNP-index = 1 if all the short readswere from KY131. The average of the SNP-index wascalculated in a 1 Mb interval using a sliding windowanalysis with 10 kb. The SNP-index graphs of the T-pool,S-pool and corresponding Δ(SNP-index) graphs wereplotted. The statistical confidence intervals of Δ(SNP-index) were calculated under the null hypothesis of noQTLs following the description of Takagi et al. (2013).

qRT-PCR analysis of the candidate genesTwo parents, DN422 and KY131, were used to iden-tify the expression patterns of the putative candidategenes. Samples were collected when they reached thepanicle initiation stage after cold water irrigation in2017. Plants with normal temperature water irrigationwere used as the control. The leaves of the two par-ents were collected at 0, 1, 3, 5, 7, 9, 11 and 13 d,for three repetitions. The samples were placed in li-quid nitrogen immediately and then stored at − 80 °Cfor total RNA isolation.Total RNA was isolated using the Trizol reagent

(Invitrogen) according to the manufacturer’s instruc-tions and was purified using the DNA-free RNA kit.The first-strand cDNA was synthesized using the

Sun et al. Rice (2018) 11:24 Page 3 of 12

Fermentas RevertAid First Strand cDNA SynthesisKit. A pair of primers was designed for each candi-date gene using the Primer Premier 5.0 software. Thehousekeeping gene Actin1 (Os05g36290) was used asthe internal control (Siahpoosh et al. 2011). Theinformation of the specific primers is provided inAdditional file 2: Table S2. The qRT-PCR reactionswere performed using a Roche LightCycler 2.10 witha 2 × SYBR Green I PCR Master Mix. The PCR reac-tion procedure was as follows: 95 °C for 5 min; 45 cy-cles of 95 °C for 10 s, 60 °C for 20 s, and 72 °C for20 s. The mean Ct values of all the biological repli-cates were normalized with the Ct values of Actin1.The relative expression level was calculated using the2-ΔΔCt method (Livak and Schmittgen 2001), whereΔΔCt = (Ct, target - Ct, actin)time x - (Ct, target - Ct,actin)time 0.

Sequencing of the nsSNP loci in RILsGenomic DNA of the 190 RILs was extracted using theCTAB method (Murray and Thompson 1980).The eightSNP loci in LOC_Os06g39750 were sequenced to iden-tify the haplotypes of RIL population. Primers for ampli-fying the SNP were designed using the Primer Premier5.0 software, and the sequences of specific primers areprovided in Additional file 2: Table S3.Total 20 μL PCR reaction mixture contained 2 μL of gen-

omic DNA (50 ng/μL), 1.5 μL of forward primer (10 μM),1.5 μL of reverse primer (10 μM), 5 μL ddH2O, and 10 μLof Pfu Master Mix (CWBio, Inc., Beijing, China) includingTaq DNA Polymerase, PCR Buffer, Mg2+ and dNTPs. ThePCR reaction was carried out in Eppendorf Mastercycler5333 thermal cycler, and the procedure was same as theabove qRT-PCR. The amplified products were checked byelectrophoresis in 1% agarose gel. PCR products were dir-ectly sequenced by the BGI Life Tech Co., Ltd.

ResultsPhenotypic characterization of the cold tolerance inDN422/KY131 RILsIn the present study, two japonica varieties, DN422 andKY131, were crossed to develop the RIL populations forthe QTL analysis and the Seq-BSA of cold tolerance.The phenotypic data were collected using the F7, F8 andF9 RILs in 2014, 2015 and 2016, continuously, and thebasic statistical analysis of the tested materials is pro-vided in Table 1. Extensive phenotyping data under coldstress and control conditions for heading time, plantheight, panicle number, and grain yield per plant wereprovide in Additional file 2: Table S4 to insights into thecold tolerance of the test materials. The mean PSSTvalues of DN422, KY131 and the RIL population in the 3years were 0.59, 0.92 and 0.76, respectively, and theirmean percent seed set values under normal water irriga-tion (PSSN) were 0.90, 0.97 and 0.89, respectively. Thecorrelation analysis showed a significantly positive cor-relation among the PSST in 2014, 2015 and 2016, re-spectively, and the PSSN had the same pattern(Additional file 2: Table S5). This indicates that the cold-water irrigation was moderate and could differentiatethe plants between cold stress and control effectively.The PSST values of the cold-tolerant parent KY131 wereextremely significant higher than those of the cold-sensitive DN422 in all 3 years (Fig. 1, Table 1), indicatingthe stronger cold tolerance of KY131 compared toDN422. Among the RIL populations, the absolute valuesof skewness and kurtosis for PSST in 3 years were allclose to 1 which indicate that the data of PSST weresuitable for the QTL analysis (Table 1) (Li et al. 2012).

SSR-based QTL mappingOne thousand SSR markers with a uniform distributionthroughout the 12 rice chromosomes were selected toscreen the polymorphism between DN422 and KY131.

Table 1 PSST and PSSN of parents and RIL population in 3 years

Traits Year DN422a KY131 RIL populationsb

Mean Range SD Skewness Kurtosis

PSST 2014 0.57 ± 0.09 0.92** ± 0.11 0.78 0.42~ 0.94 0.14 − 1.08 1.16

2015 0.61 ± 0.13 0.93** ± 0.08 0.79 0.50~ 0.98 0.12 −0.42 − 0.87

2016 0.58 ± 0.07 0.91** ± 0.14 0.70 0.31~ 0.99 0.15 0.02 −0.65

Mean 0.59 ± 0.10 0.92** ± 0.09 0.76 0.50~ 0.95 0.10 −0.10 −0.53

PSSN 2014 0.91 ± 0.05 0.97 ± 0.09 0.95 0.87~ 1.00 0.07 0.25 0.43

2015 0.88 ± 0.06 0.98 ± 0.11 0.85 0.66~ 0.99 0.03 −0.27 −0.99

2016 0.90 ± 0.14 0.96 ± 0.06 0.87 0.67~ 0.99 0.08 −0.36 −0.49

Mean 0.90 ± 0.11 0.97 ± 0.08 0.89 0.73~ 0.99 0.06 −0.07 −0.93a Means ± SD (standard deviation)b Population sample size n = 190, replications r = 2*,** significance at the level of 5 and 1%, respectively, according to Student’s t-test

Sun et al. Rice (2018) 11:24 Page 4 of 12

Among them, 158 SSR markers were polymorphic be-tween the two parents and were used to construct thegenetic linkage map of the RIL population. The map in-cluded 12 chromosomes and covered 2355.3 cM of therice genome with an average distance of 14.91 cM be-tween markers.In total, three QTLs for the PSST located on chromo-

somes 5, 6 and 7 were detected by using the ICIM mod-ule of QTL IciMapping 3.2 in 3 years (Table 2). Thetotal phenotypic variation explained (PVE) for all identi-fied QTLs ranged from 9.01 to 47.94%. qPSST5 and

qPSST7 were detected in only 1 year, and their PVE were9.01% and 14.95%, respectively. Their facticity in con-trolling the PSST of rice is suspicious. Remarkably,qPSST6 in the RM20261-RM20356 interval of chromo-some 6 was detected in all 3 years, and the PVE valuesof the 3 years were 24.30%, 47.94% and 19.11%, respect-ively, with a mean of 30.45%. In addition, their additive ef-fect values were all negative, indicating that KY131 had apositive effect on increasing the PSST. Thus, qPSST6 is amajor QTL for a high PVE and existed stably in the 3years, suggesting a key role in controlling the PSST of rice.In contrast, six QTLs with small PVE for the PSSN weredetected on chromosomes 2, 4, 6, 7, 8, and major QTLwas not found (Additional file 2: Table S6).

Sequencing of the parents and extreme poolsTwenty RILs with extreme cold tolerance (PSST rangedfrom 0.90–0.95) and 20 RILs with extreme cold sensitivity(PSST ranged from 0.50–0.63) were selected to preparethe T-pool and S-pool (Additional file 2: Table S1). Illu-mina high-throughput sequencing generated 233.62 mil-lion raw reads, and 232.30 million clean reads (99.43%)were obtained after filtering. The mean value of Q30 was91.92%, indicating that most of the bases were high qual-ity. The sequencing depths were 38-fold in DN422, 42-fold in KY131, 31-fold in the T-pool, and 45-fold in the S-pool, which could guarantee the accuracy of the BSAanalysis (Additional file 1: Figure S1, Additional file 2:Table S7). There were 336,632 SNPs between DN422 andKY131, which included 36,448 non-synonymous SNPs(nsSNP); there were 184,917 SNPs between the T-pooland the S-pool, which included 18,013 nsSNP. The rawsequencing data were deposited in NCBI (https://www.ncbi.nlm.nih.gov/), and the accession number wereSRR6327815, SRR6327816, SRR6327817, SRR6327818 forKY131, DN422, T-pool and S-pool, respectively.

ED and SNP-index analysisThe two analysis methods of Seq-BSA and the ED andSNP-index were used to identify the candidate genomeFig. 1 Panicles of DN422 and KY131 planted under cold water

irrigation condition. Scale bars, 3 cm

Table 2 QTLs for PSST of rice detected in 3 years

QTL Years Chr.a Marker interval LODb Ac PVE%d

qPSST5 2015 5 RM146-RM18332 4.69 0.04 9.01

qPSST6 2014 6 RM20261-RM20356 5.43 −0.05 24.30

2015 6 RM20261-RM20356 12.29 −0.08 47.94

2016 6 RM20261-RM20356 5.52 −0.07 19.11

qPSST7 2016 7 RM1335-RM182 3.88 0.02 14.95a Chromosomeb Logarithm of odd score for each QTLc Additive effect of the QTL. A positive value indicates that DN422 contributesthe positive allele; a negative value indicates that KY131 contributes thepositive alleled Phenotypic variation explained by A effects

Sun et al. Rice (2018) 11:24 Page 5 of 12

regions associated with PSST. The 184,917 SNPsbetween the T-pool and the S-pool were used for an as-sociation analysis through the two methods. The associ-ation threshold of the ED method was 3.82, and sixgenome regions were significantly correlated with thePSST. The result of the ED association analysis is shown inFig. 2 and Table 3. The candidate regions had a total size of23.48 Mb, distributed on chromosomes 4, 5 and 6, andcontained 4338 genes with 792 nsSNP loci among them.The SNP-index of the T-pool and S-pool were calcu-

lated for each identified SNP in the genome, and anaverage SNP-index was computed in a 1 Mb intervalusing a 10-kb sliding window. By combining the infor-mation of the SNP-index of the T-pool and S-pool, theΔ(SNP-index) was calculated, and the Δ(SNP-index)trends were visualized by means of a sliding window.Using the association threshold of 0.9532, six genomeregions distributed on chromosomes 5 and 6 that weresignificantly correlated with the PSST were identified.The total length of these regions was 3.68 Mb, and theycontained 649 genes with 93 nsSNP loci (Fig. 3, Table 4).By comparing the association results of ED and theSNP-index methods, their intersections of the genomeregions were consistent to the association regions of theSNP-index method (Table 4). The 3.68 Mb regions ofchromosome 6 by Seq-BSA was the hot region for thePSST of rice.To narrow the genome region of PSST further, map-

ping interval of the major QTL qPSST6 was comparedto the hot region of Seq-BSA. qPSST6 was detected inthe interval between RM20261-RM20356 with a geneticdistance of 28.4 cM by using the ICIM module of QTLIciMapping 3.2. RM20261 is located at 21,304,727–21,304,750 bp, and RM20356 is located at 23,654,188–23,654,207 bp on chromosome 6 of the Nipponbare

genome. qPSST6 spanned a physical distance of 2.35 Mband intersected with the hot region of Seq-BSA between21,840,000 bp and 23,654,188 bp (Fig. 4). Thus, the re-gion containing the PSST locus was narrowed to aregion of 1.81 Mb, which contains 269 predicted genes.

Identification of the candidate genes for PSSTTo identify the candidate genes for PSST, the 269predicted genes were searched in gene ontology (GO)(Ashburner et al. 2000), non-redundant protein (NR)(Deng et al. 2006), Swiss-Prot (Apweiler et al. 2004),Kyoto Encyclopedia of Genes and Genomes (KEGG)(Kanehisa et al. 2004) and Cluster of OrthologousGroups of proteins (COG) (Tatusov et al. 2000) data-bases by the BLAST software (Altschul et al. 1997). Theresults revealed that among the 269 predicted genes, 212were successfully annotated. The most enriched terms ofbiological process ontology were metabolic processes,such as carbohydrate metabolic processes (GO:0005975), DNA metabolic processes (GO:0006259), cel-lular macromolecule metabolic processes (GO:0044260),

Fig. 2 Identification of the hot-region for PSST through the ED association analysis method. X-axis represents the position of 12 chromosomes ofrice and Y-axis represents the ED. The color dots represent the ED value of every SNP locus. The black lines show the ED value of fitting results,and the red imaginary lines show the association threshold of ED

Table 3 Information of the association region by the EDassociation analysis method

Chromosomes Chromosome locations (bp) Size(Mb)

GenenumberStart End

4 5,830,000 7,910,000 2.08 324

4 20,560,000 24,640,000 4.08 770

5 26,010,000 29,950,000 3.94 873

6 2,530,000 6,440,000 3.91 756

6 19,480,000 28,910,000 9.43 1601

6 28,930,000 28,970,000 0.04 14

Total – – 23.48 4338

Sun et al. Rice (2018) 11:24 Page 6 of 12

and nucleic acid metabolic processes (GO:0090304). Themost enriched terms of molecular function werebinding, such as nucleic acid binding (GO:0003676), ironion binding (GO:0005506), and zinc ion binding, (GO:0008270). The most enriched terms of cellularcomponent were membranes, such as chloroplast outermembranes (GO:0009707), cytoplasmic membrane-bounded vesicles (GO:0016023), and plasma membranes(GO:0005886). All the annotated information is listed inAdditional file 2: Table S8. Out of the 212 annotatedgenes, only two, LOC_Os06g39740 and LOC_Os06g39750,

were annotated as the function of “response to cold (GO:0009409)”, suggesting their key roles in regulating coldtolerance in rice. Moreover, 13 predicted genes (4.83%)contained 52 nsSNP loci including three nsSNP in LOC_Os06g39750, but no non-synonymous substitution oc-curred in LOC_Os06g39740 (Additional file 2: Table S9).To determine the expression patterns of the two

genes annotated as “response to cold” in response tocold stress, a qRT-PCR was performed using the totalRNA of the two parents DN422 and KY131 subjectedto cold water conditions. Two primers pairs for theqRT-PCR analysis were designed based on the CDS se-quences of the two genes (Additional file 2: Table S2).According to the qRT-PCR results (Fig. 5), LOC_Os06g39740 was slightly induced by the cold stress after3 days in DN422 and after 1 day in KY131. The largestrelative expression quantity was nearly 2-fold at day 5in DN422 and 2.5-fold at day 7 in KY131. The expres-sion patterns of LOC_Os06g39740 have no obvious upor down-regulation for all days at the control condi-tions in DN422 and KY131. LOC_Os06g39750 exhibitsa strong inducement by cold stress at days 5, 7, 9, 11,and 13 after treatment in KY131, and the highest rela-tive expression quantity was above 9-fold at day 9.Compared to KY131, the up-regulated expression of

Fig. 3 Identification of the hot-region for PSST through the SNP-index association analysis method. X-axis represents the position of 12 chromosomesof rice and Y-axis represents the SNP-index or Δ (SNP-index). The color dots represent the SNP-index or Δ (SNP-index) value of every SNP locus. Theblack lines show the SNP-index or Δ (SNP-index) value of fitting results. a: the SNP-index graph of T-pool. b: the SNP-index graph of S-pool. c: the Δ(SNP-index) graph. The red dotted line shows the association threshold value (0.9532)

Table 4 Information of the association region by the SNP-indexassociation analysis method

Chromosomes Chromosome locations (bp) Size(Mb)

GenenumberStart End

Chr5 26,700,000 27,370,000 0.67 158

Chr6 21,840,000 22,070,000 0.23 33

Chr6 22,140,000 24,450,000 2.31 388

Chr6 24,480,000 24,700,000 0.22 27

Chr6 24,750,000 24,830,000 0.08 16

Chr6 24,980,000 25,150,000 0.17 27

Total – – 3.68 649

Sun et al. Rice (2018) 11:24 Page 7 of 12

LOC_Os06g39750 under cold stress in DN422 wereweak, and the highest relative expression quantity wasapproximately 2.6-fold at day 7. LOC_Os06g39750 ex-hibited no obvious regularity in the control conditionin both DN422 and KY131. The result of the qRT-PCRsuggested that LOC_Os06g39750 was the candidategene controlling the PSST in rice.

Among the LOC_Os06g39750, eight SNP loci wereidentified, and they were all in the CDS region of LOC_Os06g39750 (Additional file 2: Table S10). Sequencing ofthe eight SNP loci in LOC_Os06g39750 revealed thatfour haplotypes (HapI, HapII, HapIII, and HapIV)existed among the RIL population (Fig. 6, Additional file 2:Table S11). HapI, DN422 genotype, which contains 20 RILs

Fig. 4 Mapping of qPSST6 on chromosome 6 using the methods of QTL analysis and Seq-BSA. a: qPSST6 detected in the RIL population by usingthe ICIM module of QTL IciMapping 3.2. The LOD scores are shown with three distinct peaks corresponding to qPSST6 in 2014, 2015 and 2016, respectively.b: hot-region for PSST identified by SNP-index method on chromosome 6. The black line shows the Δ (SNP-index) value of fitting results. The red dottedline shows the association threshold value (0.9532), and the black arrows on the x-axis represent the location of hot-region for PSST. c: the intersection ofqPSST6 interval identified by QTL analysis and Seq-BSA. The gray region on the chromosome shows the 1.81 Mb intersection of qPSST6

Fig. 5 Expression patterns of LOC_Os06g39740 and LOC_Os06g39750 under the conditions of cold water irrigation and CK. Each line with differentpatterns on the X-axes from left to right indicate the sampling times of 0, 1, 3, 5, 7, 9, 11 and 13 day after cold water irrigating in KY131 andDN422. The Y-axes indicate the relative expression quantity. The error bars indicate SD for the data obtained in three biological replicates

Sun et al. Rice (2018) 11:24 Page 8 of 12

of S-pool and other 24 RILs. HapII, KY131 genotype, whichcontains 20 RILs of T-pool and other 9 RILs. HapIII con-tains 92 RILs, which has seven same SNPs to DN422, andonly one same SNP to KY131. HapIV contains 25 RILs,which has six same SNPs to DN422, and two same SNPs toKY131. The mean PSST of the 44 HapI and 29 HapII RILswas 0.63 and 0.91, respectively, indicating that RILs withHapII of KY131 genotype usually have a higher mean PSST.Haplotype analysis demonstrate a key role of LOC_Os06g39750 in regulating the PSST of rice once again.

DiscussionCold stress is one of the major abiotic environmentalstresses that significantly affects rice yield. Improvingthe cold tolerance of rice varieties using cold-tolerantgenes is a fast and efficient approach to reduce the yieldloss in rice cultivation regions. KY131, a japonica rice,has been widely cultivated in the northeast region ofChina for nearly 20 years. A prominent characteristic ofKY131 is the strong cold tolerance to cope with the fre-quent low temperatures that occur during the bootingstage in this high-latitude region (Yao et al. 2012; Wanget al. 2016). In this study, a RIL population derived froma cross between DN422, a cold sensitive japonica rice,and KY131 was exposed in a cold-water environmentover 3 years to acquire the PSST to map cold-tolerantgenes. The identification of cold tolerance at the bootingstage is more difficult compared to other agronomictraits in rice. Therefore, we divided the RILs into early

maturing, middle-maturing and late-maturing groups,and irrigated them independently to evaluate the PSSTof the parents and each RIL line. The criticaltemperature for the cold stress treatment at the bootingstage of rice is 17–20 °C (Zhou et al. 2010). The cold-water temperature of 17 °C selected in this study coulddifferentiate the PSST of the two parents significantly,which made the RILs vary widely (Table 1, Fig. 1). Allthe measures adopted in this study were effective atobtaining accurate PSST phenotypic data.Conventional methods of fine mapping and map-based

cloning of QTLs were very difficult for the developmentof high-density genetic map and a series of near-isogeniclines, especially the genetic population derived from ja-ponica × japonica or indica × indica crosses. For ex-ample, a long time was spent from the preliminarymapping to the cloning of the cold-tolerance gene Ctb1(Saito et al. 1995; Saito et al. 2001; Saito et al. 2010).Seq-BSA technology provides a powerful, time-savingmethod to narrow the chromosome interval harboringthe target genes/QTLs. Using the strategy of traditionalQTL mapping combined with Seq-BSA, Zheng et al.(2016) rapidly mapped and identified a novel broad-spectrum resistance gene to rice blast (Pi65(t)). ThePi65(t) region has been narrowed to 60 Kb, which con-tains 4 predicted R genes. In the present study, a majorQTL qPSST6 that explained 30.45% of the phenotypicvariation was initially mapped using the traditional QTLmapping method. Despite the relatively low density ofthe genetic map containing only 158 SSR markers, the 3years of phenotypic data of PSST demonstrate a majorQTL existing within the 28.4 cM interval between SSRmarkers RM20261-RM20356. By employing the Seq-BSA method, qPSST6 was quickly delimited to a 1.81 Mb physical interval on chromosome 6 (Fig. 4), andtwo genes, LOC_Os06g39740 and LOC_Os06g39750,were locked according to their annotated biologicalfunction as “responds to cold”. Thus, our study provideda fast and cost-effective strategy to identify the quantita-tive locus of complex trait variation. Some QTLs aboutcold tolerance at the booting stage have been reportedin recent years. These QTLs were mainly distributed onchromosome 3 (Andaya and Mackill 2003b; Suh et al.2010; Shirasawa et al. 2012; Zhu et al. 2015; Ulziibat etal. 2016), chromosome 4 (Saito et al. 2004; Saito et al.2001; Xu et al. 2008; Endo et al. 2016; Zhang et al. 2017), and chromosome 7 (Suh et al. 2010; Zhou et al. 2010).Andaya and Mackill (2003b) and Oh et al. (2004)identified a QTL controlling spikelet fertility (qCTB6)and the days to heading (dth6), respectively, locatedon chromosome 6, but the two QTLs were approxi-mately 16.4 Mb and 11.9 Mb distant to qPSST6 inthis study. Beyond those, we did not find any QTLloci distributed on chromosome 6. Thus, qPSST6 was

Fig. 6 Haplotype distribution of LOC_Os06g39750 in RIL population.Every block represents a RIL, and all RILs are arranged according totheir PSST value. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 column represent the RILswith PSST of 0.56~ 0.63, 0.63~ 0.67, 0.68~ 0.70, 0.70~ 0.72, 0.73~ 0.75,0.75~ 0.77, 0.77~ 0.82, 0.82~ 0.85, 0.85~ 0.90, 0.90~ 0.95 from topblock to bottom block, respectively. Yellow, blue, green, and redblock represent HapI, HapII, HapIII, and HapIV, respectively

Sun et al. Rice (2018) 11:24 Page 9 of 12

a new locus for cold tolerance at the booting stage inrice.Among the 269 predicted genes identified by QTL

mapping and Seq-BSA, LOC_Os06g39740 and LOC_Os06g39750 were taken into full consideration as thecandidate genes of qPSST6 according to the results ofthe GO annotation generated by Seq-BSA. The resultsof the qRT-PCR showed that LOC_Os06g39750 was in-duced strongly by cold stress in the cold-tolerant parentKY131. Haplotype analysis also demonstrated the RILswith HapII of KY131 genotype usually have a highermean PSST. These all suggesting that LOC_Os06g39750was the candidate gene of qPSST6. So far, there has notbeen any report about LOC_Os06g39750 in rice. Bysearching the Arabidopsis Information Resource (TAIR,http://www.arabidopsis.org/) database using the gene se-quence of LOC_Os06g39750, ten Arabidopsis homolo-gous genes were identified (Additional file 2: Table S12).Eight of the ten genes were the members of the 3-ketoacyl-CoA synthase family involved in the biosyn-thesis of VLCFA (very long chain fatty acids). Amongthe ten Arabidopsis homologous genes, At5g43760,At2g26640, At2g16280, At1g25450, At1g01120, andAt2g26250 were annotated as functions of “response tocold” in TAIR, indicating that they may participate theregulation of reactions to cold in Arabidopsis. An ex-pression profiling analysis revealed that most of the 3-ketoacyl-CoA synthase (KCS) family genes in Arabidop-sis, including the above six genes, responded to coldstress (Joubes et al. 2008). Therefore, it is reasonable topredict that LOC_Os06g39750 is the candidate gene forqPSST6. However, gene clone, genetic transformation,and further studies are needed to functionally validatethis conclusion.

ConclusionsThe present study using the Dongnong422/Kongyu131RIL population and a genetic linkage map containing158 SSR markers identified a novel major QTL qPSST6for cold tolerance at the booting stage of rice in the 28.4 cM region on chromosome 6. By using the combinedSeq-BSA strategy, qPSST6 was narrowed to a region of1.81 Mb, which contains 269 predicted genes. Accordingto the results of the gene ontology (GO) analysis, LOC_Os06g39740 and LOC_Os06g39750 were annotated as“response to cold”. LOC_Os06g39750 exhibited strongup-regulated expressions at the 5, 7, 9, 11, and 13 dayafter the cold-water treatment in KY131. Haplotype ana-lysis also demonstrate a key role of LOC_Os06g39750 inregulating the PSST of rice, suggesting that it was thecandidate gene of qPSST6. This study provided a fastand cost-effective strategy to identify cold-tolerant genesat the booting stage in rice.

Additional files

Additional file 1: Figure S1. Distributions of coverage depth onchromosomes of the sequencing samples. X-axe indicates the 12 chro-mosomes of rice. Y-axe indicates the log2 value of coverage depth corre-sponding to chromosome location. (DOC 1130 kb)

Additional file 2: Table S1. The PSST of T-pool and S-pool. Table S2.Sequences of specific primers for PSST candidate genes. Table S3. Se-quences of specific primers for SNP loci in LOC_Os06g39750. Table S4.Heading time, plant height, panicle number, and grain yield per plant ofparents and RIL population in 2014, 2015 and 2016. Table S5. Coeffi-cients of correlation (r) among PSST and PSSN in 2014, 2015 and 2016.Table S6. QTLs for PSSN of rice detected in 3 years. Table S7. Summaryof Seq-BSA data for each sample. Table S8. Annotation information ofpredicted genes for PSST. Table S9. nsSNPs between DN422 and KY131along with their S-pool and T-pool. Table S10. Information of SNP/InDelin LOC_Os06g39740 and LOC_Os06g39750. Table S11. Haplotype ofLOC_Os06g39750 in RIL population and mean PSST of each RIL. Table S12.The Arabidopsis homologous genes of LOC_Os06g39750. (XLS 174 kb)

AbbreviationsBSA: Bulked-segregant analysis; COG: Cluster of orthologous groups ofproteins; ED: Euclidean distance; GO: Gene ontology; KEGG: KyotoEncyclopedia of Genes and Genomes; MAS: Marker assisted selection;NGS: Next-generation sequencing; NR: Non-redundant protein; PSSN: Percentseed set under normal water irrigation; PSST: Percent seed set under coldwater treatment; PVE: Phenotypic variation explained; QTL: Quantitative traitloci; RIL: Recombinant inbred line; S-pool: Sensitive pool; SSD: Single seeddescent; T-pool: Tolerant pool

AcknowledgmentsWe would like to thank the “Young Talents” Project of Northeast AgriculturalUniversity for its research funding support, and the National InnovationCenter of Regional Technology in Rice for the laboratory and field support.

FundingThis work was supported by the “Young Talents” Project of NortheastAgricultural University (16QC03), National Natural Science Foundation(31701507), Youth Science Foundation of Heilongjiang Province(QC2017015), Major Scientific and Technological Bidding Projects ofHeilongjiang Province (GA14B102–02).

Availability of data and materialsAll relevant data are provided in Tables and in Supplementary Tables. Theraw sequencing data are deposited in NCBI (https://www.ncbi.nlm.nih.gov/).

Authors’ contributionsSJ and ZDT designed research; YLM performed QTL mapping; WJG analyzedBSA sequence data; LHL, ZHL, and JY irrigated the cold water and got thephenotyping data; XDW and ZMH performed qRT-PCR analysis, FMF analyzedcandidate genes; ZHW prepared the T-pool and S-pool; SJ wrote manuscript;ZDT corrected manuscript. All authors have read and approved themanuscript.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Author details1College of Agriculture, Northeast Agricultural University, Harbin 150030,China. 2The Institute of Industrial Crops of Heilongjiang Academy ofAgricultural Sciences, Harbin 150086, China. 3College of Life Science,Northeast Agricultural University, Harbin 150030, China.

Sun et al. Rice (2018) 11:24 Page 10 of 12

Received: 18 August 2017 Accepted: 4 April 2018

ReferencesAbe A, Kosugi S, Yoshida K, natsume S, Takagi H, Kanzaki H, Matsumura H,

Yoshida K, Mitsuoka C, Tamiru M (2012) Genome sequencing revealsagronomically important loci in rice using MutMap. Nat Biotech 30:174–178

Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ(1997) Gapped BLAST and PSI-BLAST: a new generation of protein databasesearch programs. Nucleic Acids Res 25(17):3389–3402

Andaya V, Mackill D (2003b) QTLs conferring cold tolerance at the booting stageof rice using recombinant inbred lines from a japonica × indica cross. TheorAppl Genet 106(6):1084–1090

Andaya VC, Mackill DJ (2003a) Mapping of QTLs associated with cold toleranceduring the vegetative stage in rice. J Exp Bot 54(392):2579

Apweiler R, Bairoch A, Wu CH, Barker WC, Boeckmann B, Ferro S, Gasteiger E,Huang HZ, Lopez R, Magrane M, Martin MJ, Natale DA, O’Donovan C,Redaschi N, Yeh SL (2004) UniProt: the universal protein knowledgebase.Nucleic Acids Res 32:115–119

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, DolinskiK, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S,Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G (2000) Geneontology: tool for the unification of biology. Nat Genet 25(1):25–29

Chen X, Temnykh S, Xu Y, Cho YG, McCouch SR (1997) Development of amicrosatellite framework map providing genome-wide coverage in rice(Oryza sativa L.). Theor Appl Genet 95(4):553–567

Deng YY, Li JQ, Wu SF, Zhu YP, Chen YW, He FC (2006) Integrated nr database inprotein annotation system and its localization. Comput Eng 32(5):71–74

Endo T, Chiba B, Wagatsuma K, Saeki K, Ando T, Shomura A, Mizubayashi T, Ueda T,Yamamoto T, Nishio T (2016) Detection of QTLs for cold tolerance of ricecultivar “Kuchum” and effect of QTL pyramiding. Theor Appl Genet 129(3):631

Fekih R, Takagi H, Tamiru M, Abe A, Natsume S, Yaegashi H, Sharma S, Sharma S,Kanzaki H, Matsumura H, Saitoh H, Mitsuoka C, Utsushi H, Saitoh H (2013)MutMap+: genetic mapping and mutant identification without crossing inrice. PLoS One 8(7):e68529

Han YC, Lv P, Hou SL, Li SY, Ji GS, Ma X, Du RH, Liu GQ (2015) Combining nextgeneration sequencing with bulked segregant analysis to fine map a stemmoisture locus in Sorghum (Sorghum bicolor L. Moench). PLoS One 10(5):e0127065

Hill JT, Demarest BL, Bisgrove BW, Gorsi B, Su YC, Yost HJ (2013) MMAPPR:mutation mapping analysis pipeline for pooled RNA-seq. Genome Res 23(4):687–697

International Rice Genome Sequencing Project (IRGSP) (2005) The map-basedsequence of the rice genome. Nature 436:793–800

Jena KK, Kim SM, Suh JP, Yang CI, Kim YJ (2012) Identification of cold-tolerantbreeding lines by quantitative trait loci associated with cold tolerance in rice.Crop Sci 51(2):517–523

Jia Y, Zou DT, Wang JG, Liu HL, Inayat MA, Sha HJ, Zheng HL, Sun J, Zhao HW (2015)Effect of low water temperature at reproductive stage on yield and glutamatemetabolism of rice (Oryza sativa L.) in China. Field Crops Res 175:16–25

Joubes J, Raffaele S, Bourdenx B, Garcia C, Laroche-Traineau J, Moreau P,Domergue F, Lessire R (2008) The VLCFA elongase gene family inArabidopsis thaliana: phylogenetic analysis, 3D modelling and expressionprofiling. Plant Mol Biol 67:547–566

Kanehisa M, Goto S, Kawashima S, Okuno Y, Hattori AM (2004) The KEGGresource for deciphering the genome. Nucleic Acids Res 32:277–280

Li H, Durbin R (2009) Fast and accurate short read alignment with burrows-wheeler transform. Bioinformatics 25:1754–1760

Li SH, Wang CS, Chang XP, Jing RL (2012) Genetic dissection of developmentalbehavior of grain weight in wheat under diverse temperature and waterregimes. Genetica 140:393–405

Livaja M, Wang Y, Wieckhorst S, Haseneyer G, Seidel M, Hahn V, Knapp SJ, Taudien S,Schön CC, Bauer E (2013) BSTA: a targeted approach combines bulkedsegregant analysis with next-generation sequencing and de novotranscriptome assembly for SNP discovery in sunflower. BMC Genomics 14:628

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data usingreal-time quantitative pcr and the 2-ΔΔCt method. Methods 25:402–408

Matsunaga K (2005) Establishment of an evaluation method for cold tolerance atthe booting stage of rice using deep water irrigation system anddevelopment of highly cold-tolerant rice varieties by combining coldtolerance genes. Bull Miyagi Furukawa Agric Exp Stn 4:1–78

McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, GarimellaK, Altshuler D, Gabriel S, Daly M, DePristo MA (2010) The genome analysistoolkit: a MapReduce framework for analyzing next-generation DNAsequencing data. Genome Res 20(9):1297–1303

Michelmore RW, Paran I, Kesseli RV (1991) Identification of markers linked todisease-resistance genes by bulked segregant analysis: a rapid method todetect markers in specific genomic regions by using segregatingpopulations. Proc Natl Acad Sci U S A 88:9828–9832

Murray MG, Thompson WF (1980) Rapid isolation of high molecular weight plantDNA. Nucl Acids Res 8:4321–4325

Oh CS, Choi YH, Lee SJ, Yoon DB, Moon HP, Ahn SN (2004) Mapping of quantitativetrait loci for cold tolerance in weedy rice. Breeding Sci 54(4):373–380

Saito K, Hayano Y, Kuroki M, Sato Y (2010) Map-based cloning of the rice coldtolerance gene Ctb1. Plant Sci 179:97–102

Saito K, Hayano-Saito Y, Maruyama-Funatsuki W, Sato Y, Kato A (2004) Physicalmapping and putative candidate gene identification of a quantitative traitlocus Ctb1 for cold tolerance at the booting stage of rice. Theor Appl Genet109(3):515–522

Saito K, Miura K, Nagano K, Hayano-Saito Y, Araki H, Kato A (2001) Identificationof two closely linked quantitative trait loci for cold tolerance onchromosome 4 of rice and their association with anther length. Theor ApplGenet 103(6):862–868

Saito K, Miura K, Nagano K, Hayano-Saito Y, Saito A, Araki H, Kato A (1995)Chromosomal location of quantitative trait loci for cool tolerance at thebooting stage in rice variety “Norin-PL8”. Japanese J Breeding 45(3):337–340

Salvi S, Tuberosa R (2005) To clone or not to clone plant QTLs: present andfuture challenges. Trends Plant Sci 10:297–304

Satake T, Hayase H (1970) Male sterility caused by cooling treatment at theyoung micro-spore stage in rice plants. V. Estimations of pollendevelopmental stage and the most sensitive stage to coolness. Proc Crop SciSoc Jpn 39(4):468–473

Schneeberger K, Ossowski S, Lanz C, Juul T, Petersen AH, Nielsen KL, JorgensenJE, Weigel D, Andersen SU (2009) SHOREmap: simultaneous mapping andmutation identification by deep sequencing. Nat Methods 6:550–551

Shirasawa S, Endo T, Nakagomi K, Yamaguchi M, Nishio T (2012) Delimitation of aQTL region controlling cold tolerance at booting stage of a cultivar,‘Lijiangxintuanheigu’, in rice, Oryza sativa L. Theor Appl Genet 124(5):937–946

Siahpoosh MR, Sanchez DH, Schlereth A, Scofield GN, Furbank RT, Dongen JT,Kopka J (2011) Modification of OsSUT1 gene expression modulates the saltresponse of rice Oryza sativa cv. Taipei 309. Plant Sci 182:101–111

Singh VK, Khan AW, Saxena RK, Kumar V, Kale SM, Sinha P, Chitikineni A,Pazhamala LT, Garg V, Sharma M, Kumar CVS, Parupalli S, Vechalapu S, Patil S,Muniswamy S, Ghanta A, Yamini KN, Dharmaraj PS, Varshney RK (2016) Next-generation sequencing for identification of candidate genes for fusarium wiltand sterility mosaic disease in pigeonpea (Cajanus cajan). Plant Biotech J14(5):1183–1194

Song J, Li Z, Liu Z, Guo Y, Qiu LJ (2017) Next-generation sequencing frombulked-segregant analysis accelerates the simultaneous identification of twoqualitative genes in soybean. Frontiers Plant Sci 8. https://doi.org/10.3389/fpls.2017.00919

Suh JP, Jeung JU, Lee JI, Choi YH, Yea JD, Virk PS, Mackill DJ, Jena KK (2010)Identification and analysis of QTLs controlling cold tolerance at thereproductive stage and validation of effective QTLs in cold-tolerantgenotypes of rice (Oryza sativa L.). Theor Appl Genet 120:985–995

Takagi H, Abe A, Yoshida K, Kosugi S, Natsume S, Mitsuoka C, Uemura A, Utsushi H,Tamiru M, Takuno S, Innan H, Cano LM, Kamoun S, Terauchi R (2013) QTL-seq:rapid mapping of quantitative trait loci in rice by whole genome resequencingof DNA from two bulked populations. The. Plant J 74(1):174–183

Tatusov RL, Galperin MY, Natale DA, Koonin EV (2000) The COG database: a toolfor genome scale analysis of protein functions and evolution. Nucleic AcidsRes 28:33–36

Trick M, Adamski NM, Mugford SG, Jiang CC, Febrer M, Uauy C (2012) CombiningSNP discovery from next-generation sequencing data with bulked segregantanalysis (BSA) to fine-map genes in polyploid wheat. BMC Plant Biol 12:14

Ulziibat B, Ohta H, Fukushima A, Shirasawa S, Kitashiba H, Nishio T (2016)Examination of candidates for the gene of cold tolerance at the bootingstage in a delimited QTL region in rice cultivar ‘Lijiangxintuanheigu’.Euphytica 211(3):331–341

Wambugu P, Ndjiondjop MN, Furtado A, Henry R (2017) Sequencing of bulks ofsegregants allows dissection of genetic control of amylose content in rice.Plant Biotech J. https://doi.org/10.1111/pbi.12752

Sun et al. Rice (2018) 11:24 Page 11 of 12

Wang D, Liu JL, Li CG, Kang HX, Wang Y, Tan XQ, Liu MH, Deng YF, Wang ZL, LiuY, Zhang DY, Xiao YH, Wang GL (2016) Genome-wide association mappingof cold tolerance genes at the seedling stage in rice. Rice 9(1):61

Wang LZ, Wang LM, Xiang HT, Luo Y, Li R, Li ZJ, Wang CY, Meng Y (2016)Relationship of photosynthetic efficiency and seed-setting rate in twocontrasting rice cultivars under chilling stress. Photosynthetica 54(4):581–588

Wenger WJ, Schwartz K, Sherlock G (2010) Bulk segregant analysis by high-throughput sequencing reveals a novel xylose utilization gene fromSaccharomyces cerevisiae. PLoS Genet 6:e1000942

Xu LM, Zhou L, Zeng YW, Wang FM, Zhang HL, Shen SQ, Li ZC (2008)Identification and mapping of quantitative trait loci for cold tolerance at thebooting stage in a japonica rice near-isogenic line. Plant Sci 174(3):340–347

Yao Y, Miao Y, Huang S, Gao L, Ma X, Zhao G, Gnyp ML (2012) Active canopysensor-based precision N management strategy for rice. Agron Sustain Dev32(4):925–933

Zhang Q, Chen QH, Wang SL, Hong YH, Wang ZL (2014) Rice and cold stress:methods for its evaluation and summary of cold tolerance-relatedquantitative trait loci. Rice 7:24

Zhang ZH, Su L, Li W, Chen W, Zhu YG (2005) A major QTL conferring coldtolerance at the early seedling stage using recombinant inbred lines of rice(Oryza sativa L.). Plant Sci 168(2):527–534

Zhang ZY, Li JJ, Pan YH, Li JL, Zhou L, Shi HL, Zeng YW, Guo HF, Yang SM, ZhengWW, Yu JP, Sun XM, Li GL, Ding YL, Ma L, Shen SQ, Dai LY, Zhang HL, YangSH, Guo Y, Li ZC (2017) Natural variation in CTB4a enhances rice adaptationto cold habitats. Nat Commun 8:14788

Zheng W, Wang Y, Wang L, Ma ZB, Zhao JM, Wang P, Zhang LX, Liu ZH, Lu XC(2016) Genetic mapping and molecular marker development for Pi65(t), anovel broad-spectrum resistance gene to rice blast using next-generationsequencing. Theor Appl Genet 129(5):1035

Zhou L, Zeng YW, Zheng WW, Tang B, Yang SM, Zhang HL, Li JJ, Li ZC (2010)Fine mapping a QTL qCTB7 for cold tolerance at the booting stage on ricechromosome 7 using a near-isogenic line. Theor Appl Genet 121(5):895

Zhu Y, Chen K, Mi X, Chen T, Ali J, Ye G, Xu JL, Li ZK (2015) Identification and finemapping of a stably expressed QTL for cold tolerance at the booting stage usingan interconnected breeding population in rice. PLoS One 10(12):e0145704

Sun et al. Rice (2018) 11:24 Page 12 of 12